{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UF4XPNZ4EE2K3Q35ABIIHPAOL3","short_pith_number":"pith:UF4XPNZ4","canonical_record":{"source":{"id":"2607.12963","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T17:01:12Z","cross_cats_sorted":[],"title_canon_sha256":"9f226e1e14b7c95386f1cd547a5d467da6823585cb5c11df9af802a6a0b231ab","abstract_canon_sha256":"9e5db9e0712eac2d75c6ad2743ee772c8563811ee1c897bc4cabcc5f8854efd3"},"schema_version":"1.0"},"canonical_sha256":"a17977b73c2134adc37d005083bc0e5efea3e6221d22b9e9ea4ba8fc1103af52","source":{"kind":"arxiv","id":"2607.12963","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.12963","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"arxiv_version","alias_value":"2607.12963v1","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.12963","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"pith_short_12","alias_value":"UF4XPNZ4EE2K","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"pith_short_16","alias_value":"UF4XPNZ4EE2K3Q35","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"pith_short_8","alias_value":"UF4XPNZ4","created_at":"2026-07-15T01:22:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UF4XPNZ4EE2K3Q35ABIIHPAOL3","target":"record","payload":{"canonical_record":{"source":{"id":"2607.12963","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T17:01:12Z","cross_cats_sorted":[],"title_canon_sha256":"9f226e1e14b7c95386f1cd547a5d467da6823585cb5c11df9af802a6a0b231ab","abstract_canon_sha256":"9e5db9e0712eac2d75c6ad2743ee772c8563811ee1c897bc4cabcc5f8854efd3"},"schema_version":"1.0"},"canonical_sha256":"a17977b73c2134adc37d005083bc0e5efea3e6221d22b9e9ea4ba8fc1103af52","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-15T01:22:28.042295Z","signature_b64":"mx6eqwUQQ7g8d8BIRyjIhSimEAH7V5jYaWR5uo90ziEIA8HcJaFOQ7S3L48XB8/+yVqxGZJ5ixTrCXK/k591BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a17977b73c2134adc37d005083bc0e5efea3e6221d22b9e9ea4ba8fc1103af52","last_reissued_at":"2026-07-15T01:22:28.041395Z","signature_status":"signed_v1","first_computed_at":"2026-07-15T01:22:28.041395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.12963","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-15T01:22:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8cOVEbMjwoJ8piEH9fua2Uv5MahoMhXnhpk0qhHOouwP2601rcroseuEDeRjJ9o+xsDdYK6DN18vXOrkB5E6CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T15:18:37.730853Z"},"content_sha256":"e551427f49674de4b0ba936b68042c8be080acd60c350cc986a75bebb4d343b7","schema_version":"1.0","event_id":"sha256:e551427f49674de4b0ba936b68042c8be080acd60c350cc986a75bebb4d343b7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UF4XPNZ4EE2K3Q35ABIIHPAOL3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Illusion of Robustness: Aggregate Accuracy Hides Prediction Flips under Task-Irrelevant Context","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Diyi Yang, Sanmi Koyejo, Yanzhe Zhang","submitted_at":"2026-07-14T17:01:12Z","abstract_excerpt":"As large language models (LLMs) grow more capable, they are increasingly deployed in context-rich settings where task inputs are often accompanied by long, partially irrelevant context. In a controlled setting, we find that state-of-the-art models often appear robust to task-irrelevant context at the aggregate level: prepending it to benchmark questions causes little change in overall accuracy. This aggregate stability, however, masks significant per-example instability. Even semantically meaningless pseudo-words, formed by randomly combining characters, can markedly shift model predictions on"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.12963","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2607.12963/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-15T01:22:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3Jgil50vMYgu+L0MZgKmarKdb7Mxl+7Q4eaI81gV864cEf4wq5L4Hqdn7yNPJp5kx6wDjZPvqpdD2ptaZ8XXDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T15:18:37.731378Z"},"content_sha256":"9e0ab215a70357c8a60e155527da47b8dec959f5b94a59f6dd94e10019e5b807","schema_version":"1.0","event_id":"sha256:9e0ab215a70357c8a60e155527da47b8dec959f5b94a59f6dd94e10019e5b807"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UF4XPNZ4EE2K3Q35ABIIHPAOL3/bundle.json","state_url":"https://pith.science/pith/UF4XPNZ4EE2K3Q35ABIIHPAOL3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UF4XPNZ4EE2K3Q35ABIIHPAOL3/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-17T15:18:37Z","links":{"resolver":"https://pith.science/pith/UF4XPNZ4EE2K3Q35ABIIHPAOL3","bundle":"https://pith.science/pith/UF4XPNZ4EE2K3Q35ABIIHPAOL3/bundle.json","state":"https://pith.science/pith/UF4XPNZ4EE2K3Q35ABIIHPAOL3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UF4XPNZ4EE2K3Q35ABIIHPAOL3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UF4XPNZ4EE2K3Q35ABIIHPAOL3","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9e5db9e0712eac2d75c6ad2743ee772c8563811ee1c897bc4cabcc5f8854efd3","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T17:01:12Z","title_canon_sha256":"9f226e1e14b7c95386f1cd547a5d467da6823585cb5c11df9af802a6a0b231ab"},"schema_version":"1.0","source":{"id":"2607.12963","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.12963","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"arxiv_version","alias_value":"2607.12963v1","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.12963","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"pith_short_12","alias_value":"UF4XPNZ4EE2K","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"pith_short_16","alias_value":"UF4XPNZ4EE2K3Q35","created_at":"2026-07-15T01:22:28Z"},{"alias_kind":"pith_short_8","alias_value":"UF4XPNZ4","created_at":"2026-07-15T01:22:28Z"}],"graph_snapshots":[{"event_id":"sha256:9e0ab215a70357c8a60e155527da47b8dec959f5b94a59f6dd94e10019e5b807","target":"graph","created_at":"2026-07-15T01:22:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.12963/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) grow more capable, they are increasingly deployed in context-rich settings where task inputs are often accompanied by long, partially irrelevant context. In a controlled setting, we find that state-of-the-art models often appear robust to task-irrelevant context at the aggregate level: prepending it to benchmark questions causes little change in overall accuracy. This aggregate stability, however, masks significant per-example instability. Even semantically meaningless pseudo-words, formed by randomly combining characters, can markedly shift model predictions on","authors_text":"Diyi Yang, Sanmi Koyejo, Yanzhe Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T17:01:12Z","title":"The Illusion of Robustness: Aggregate Accuracy Hides Prediction Flips under Task-Irrelevant Context"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.12963","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e551427f49674de4b0ba936b68042c8be080acd60c350cc986a75bebb4d343b7","target":"record","created_at":"2026-07-15T01:22:28Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"9e5db9e0712eac2d75c6ad2743ee772c8563811ee1c897bc4cabcc5f8854efd3","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-07-14T17:01:12Z","title_canon_sha256":"9f226e1e14b7c95386f1cd547a5d467da6823585cb5c11df9af802a6a0b231ab"},"schema_version":"1.0","source":{"id":"2607.12963","kind":"arxiv","version":1}},"canonical_sha256":"a17977b73c2134adc37d005083bc0e5efea3e6221d22b9e9ea4ba8fc1103af52","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a17977b73c2134adc37d005083bc0e5efea3e6221d22b9e9ea4ba8fc1103af52","first_computed_at":"2026-07-15T01:22:28.041395Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-15T01:22:28.041395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mx6eqwUQQ7g8d8BIRyjIhSimEAH7V5jYaWR5uo90ziEIA8HcJaFOQ7S3L48XB8/+yVqxGZJ5ixTrCXK/k591BQ==","signature_status":"signed_v1","signed_at":"2026-07-15T01:22:28.042295Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.12963","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e551427f49674de4b0ba936b68042c8be080acd60c350cc986a75bebb4d343b7","sha256:9e0ab215a70357c8a60e155527da47b8dec959f5b94a59f6dd94e10019e5b807"],"state_sha256":"7ef28f1bc966a129a008cbb50e88fd31399a94bf6592a8615634624b16a93f51"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VlRN+A7VOxWAeaBfhIQ/TUwf7zRivN+AwElTZIU0g07G7xX3E3tOZYDybAKJxw3vrTXfhxxKiY1e216GcUADBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T15:18:37.737778Z","bundle_sha256":"835663e8998190f08aa5d7c68824b00a79b8d8ad0baab8f2282e9a4e7c1974d7"}}